PreProccesing and Analysis

---

zomato <- read.csv("C:/Users/Rajnandini/Desktop/ds cp/sycsd grp 5/zomato.csv", encoding = "latin1")

# Visualize the percentage of data from different countries
library(ggplot2)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(scales)

countries <- unique(zomato$Country.Code)
countries

##  [1] 162  30 216  14  37 184 214   1  94 148 215 166 189 191 208

country_counts <- sapply(countries, function(c) sum(zomato$Country.Code == c))
country_counts

##  [1]   22   60  434   24    4   20   60 8652   21   40   80   20   60   20   34

country_data <- data.frame(Country = countries, Count = country_counts)
country_data

ggplot(country_data, aes(x = "", y = Count, fill = Country)) +
  geom_bar(width = 1, stat = "identity") +
  coord_polar("y", start = 0) +
  theme_void() +
  theme(legend.position = "top")

# Filter data for India with country code 1
zomato_india <- zomato[zomato$Country.Code == 1, ]
zomato_india

# Clean up data
zomato_india$Has.Table.booking <- ifelse(zomato_india$Has.Table.booking == "Yes", 1, 0)
zomato_india$Has.Online.delivery <- ifelse(zomato_india$Has.Online.delivery == "Yes", 1, 0)
zomato_india$Is.delivering.now <- ifelse(zomato_india$Is.delivering.now == "Yes", 1, 0)
zomato_india$Rating.text <- as.factor(zomato_india$Rating.text)
zomato_india <- zomato_india[zomato_india$Rating.text != "Not rated", ]

# Calculate mean for Average Cost for two
average_cost_mean <- mean(zomato_india$Average.Cost.for.two[zomato_india$Average.Cost.for.two != 0])
zomato_india$Average.Cost.for.two[zomato_india$Average.Cost.for.two == 0] <- average_cost_mean

# Visualize cities in India
city_counts <- zomato_india %>% group_by(City) %>% summarize(Count = n())
ggplot(city_counts, aes(x = "", y = Count, fill = City)) +
  geom_bar(width = 1, stat = "identity") +
  coord_polar("y", start = 0) +
  theme_void() +
  theme(legend.position = "bottom")

# Visualize rating vs number of votes
ggplot(zomato_india, aes(x = Aggregate.rating, y = Votes)) +
  geom_bar(stat = "identity") +
  labs(x = "Rating", y = "Number of Votes") +
  ggtitle("Rating vs Number of Votes")

# Visualize rating vs average cost for two
ggplot(zomato_india, aes(x = Aggregate.rating, y = Average.Cost.for.two)) +
  geom_bar(stat = "identity", fill = "green") +
  labs(x = "Rating", y = "Average Cost for Two (in Indian Rupees)") +
  ggtitle("Rating vs Avg. Cost for Two")

ggplot(zomato_india, aes(x = Aggregate.rating, y = Average.Cost.for.two)) +
  geom_bar(stat = "identity", fill = "green") +
  scale_y_continuous(labels = comma) +  # Format y-axis labels with commas
  labs(x = "Rating", y = "Average Cost for Two (in Indian Rupees)") +
  ggtitle("Rating vs Avg. Cost for Two")

zomato_india$Cuisine.code <- gsub(",.*", "", zomato_india$Cuisines)
zomato_india

labels <- c("Afghani", "American", "Andhra", "Arabian", "Asian", "Assamese", "Awadhi", "Bakery", "Bengali", "Beverages",
            "Bihari", "Biryani", "British", "Burger", "Burmese", "Cafe", "Charcoal Grill", "Chinese", "Continental", 
            "Desserts", "Drinks Only", "European", "Fast Food", "Finger Food", "French", "Goan", "Greek", "Gujarati", 
            "Healthy Food", "Hyderabadi", "Ice Cream", "Indian", "Italian", "Japanese", "Juices", "Kashmiri", "Kerala", 
            "Korean", "Lebanese", "Lucknowi", "Maharashtrian", "Malaysian", "Malwani", "Mediterranean", "Mexican", 
            "Middle Eastern", "Mithai", "Modern Indian", "Mughlai", "Naga", "Nepalese", "North Eastern", "North Indian", 
            "Oriya", "Parsi", "Persian", "Pizza", "Portuguese", "Rajasthani", "Raw Meats", "Salad", "Seafood", 
            "South American", "South Indian", "Spanish", "Steak", "Street Food", "Sushi", "Tea", "Tex-Mex", "Thai", 
            "Tibetan", "Turkish", "Vietnamese")

values <- seq_along(labels) - 1  # Assign values 0 to n-1

# Remove duplicate column names if any
zomato_india <- zomato_india[, !duplicated(names(zomato_india))]

# Modify the "Cuisine" column based on labels and values, with a default value of -1 for unmatched values
zomato_india <- zomato_india %>%
  mutate(Cuisine.code = recode(Cuisine.code, !!!setNames(values, labels), .default = -1))

zomato_india

# Drop unwanted columns and reorder
# Drop unwanted columns and reorder
zomato_india_cleaned <- zomato_india[, !names(zomato_india) %in% c("Cuisines", "Country.Code", "Rating.color", "Switch.to.order.menu", "Currency", "Address", "Locality", "Locality.Verbose")]
zomato_india_cleaned

# Select columns for zomato_india_cleaned
# Extract City column from the original dataset
city_column <- zomato_india$City

# Reorder the columns to place the "City" column in the desired position
zomato_india_cleaned <- zomato_india_cleaned[, c("Restaurant.Name", "City", "Average.Cost.for.two", "Has.Table.booking", "Has.Online.delivery", "Is.delivering.now", "Votes", "Aggregate.rating", "Rating.text", "Cuisine.code")]

# Write the cleaned dataset to a new CSV file
write.csv(zomato_india_cleaned, file = "/Users/Rajnandini/Desktop/ds cp/sycsd grp 5/zomatoIndiaCleaned.csv", row.names = FALSE)
zomato_india_cleaned